548 EIGHTH PACIFIC SCIENCE CONGRESS 



wind-driven water masses piled up against the land barriers. As a mat- 

 ter of fact, Ekman's theory assumes no boundaries and a constant lati- 

 tude. We can prove the existence of slope current in an ocean having 

 boundaries partly or completely enclosing it. The slope current is 

 uniform from the surface down to the bottom. This fact seems in 

 favor of the theoretical result we have obtained. Still, we must give 

 attention to the fact that the velocity of slope current is always inversely 

 proportional to the depth of the sea. When the depth is large as we see 

 in the actual oceans, the slope current will not be strong enough to 

 account for those large velocities we have obtained at the depth twice 

 or three times as large as D^. 



We don't know an appropriate explanation of the theoretical re- 

 sult that the effect of winds can be felt at a depth several times as large 

 as Ekman's depth of frictional influence. It would be hoped someone 

 may be able to solve this question satisfactorily in the near future. 



15. Summary 



(1) A theory of the general circulation of water in the Pacific Ocean 

 produced by the semi-permanent wind system prevailing over this ocean 

 is propounded. 



(2) The velocity is used to express the water motion which has 

 formerly been explained by several authors in terms of mass transport. 



(3) The Pacific Ocean is considered to be a rectangular ocean ex- 

 tending from 60° S to 60^ N latitudes and from 0° to 120'' longitude, 

 and a zonal distribution of the wind system determined from actual 

 observations has been assumed. 



(4) The effects of horizontal turbulence and the meridional varia- 

 tion of the Coriolis forces have been taken into account. 



(5) The patterns of horizontal circulation are obtained in terms 

 of streamlines for the sea surface and several deeper layers specified by 

 the ratio zjD^ where z is the geometrical depth below the surface and 

 D^ the depth of the frictional influence, a measure of vertical turbulence. 



(6) Surface circulation has a pattern similar to that actually ob- 

 served and does not differ much from Munk's result obtained in terms 

 of mass transport. We have very strong western currents and boundary 

 vortices. 



(7) The magnitude of the Kuroshio and other western currents 

 was computed from the distribution of the streamlines in each level. 

 The velocity is inversely proportional to D^ so that we can determine 

 it by assuming an appropriate value for D^. A value of D.^ between 

 50 m and 150 m gives values most reasonable and consistent with the 

 actual observations. 



